A machine learning framework for the prediction of chromatin folding in Drosophila using epigenetic features

• Published in: PeerJ. Computer science 2020-11, Vol.6, p.e307-e307, Article e307
• Main Authors: Rozenwald, Michal B, Galitsyna, Aleksandra A, Sapunov, Grigory V, Khrameeva, Ekaterina E, Gelfand, Mikhail S
• Format: Article
• Language: English
• Subjects: Analysis; Artificial neural networks; Binding sites; Bioinformatics; Biotechnology; Chromatin; Chromosomes; Computational Biology; Data Mining and Machine Learning; Data Science; Datasets; Deoxyribonucleic acid; DNA; DNA binding proteins; Domains; Drosophila; Dynamic programming; Epigenetic inheritance; Epigenetics; Folding; Fruit flies; Gene expression; Generalized linear models; Genes; Genomes; Innovations; Insects; Machine learning; Mammals; Molecular Biology; Neural networks; Protein binding; Proteins; Recurrent neural networks; Regression models; Regularization; RNA polymerase
• ISSN: 2376-5992; 2376-5992
• DOI: 10.7717/PEERJ-CS.307

Technological advances have lead to the creation of large epigenetic datasets, including information about DNA binding proteins and DNA spatial structure. Hi-C experiments have revealed that chromosomes are subdivided into sets of self-interacting domains called Topologically Associating Domains (TADs). TADs are involved in the regulation of gene expression activity, but the mechanisms of their formation are not yet fully understood. Here, we focus on machine learning methods to characterize DNA folding patterns in based on chromatin marks across three cell lines. We present linear regression models with four types of regularization, gradient boosting, and recurrent neural networks (RNN) as tools to study chromatin folding characteristics associated with TADs given epigenetic chromatin immunoprecipitation data. The bidirectional long short-term memory RNN architecture produced the best prediction scores and identified biologically relevant features. Distribution of protein Chriz (Chromator) and histone modification H3K4me3 were selected as the most informative features for the prediction of TADs characteristics. This approach may be adapted to any similar biological dataset of chromatin features across various cell lines and species. The code for the implemented pipeline, Hi-ChiP-ML, is publicly available: https://github.com/MichalRozenwald/Hi-ChIP-ML.